To evaluate the clinical utility of percutaneous drainage of pancreatic fistula following pancreatectomy with real-time CT-fluoroscopic guidance.

During January 2007 through March 2013, of 295 patients who underwent pancreatectomy, 20 patients received percutaneous drainage of pancreatic fistula with real-time CT-fluoroscopic guidance. The mean diameter of pancreatic fluid collections was 8.1±2.7 (SD)cm (range: 3.5–15.0cm). Feasibility, safety, and clinical success were evaluated. Primary and secondary clinical successes were defined respectively as the resolution of pancreatic fistula by initial drainage alone, and after additional intervention. Factors affecting primary clinical success and the drainage period were also evaluated.

Drainage catheters were placed in planned sites in all patients. No major complication occurred except in 1/20 patient (5%) who experienced endotoxin shock. Primary and secondary clinical success rates were, respectively, 50% (10/20) and 90% (18/20). An amylase level greater than 30,000IU/L in the fluid collection was a significant factor lowering the primary clinical success rate (P<0.02) and prolonging the drainage period (>30 days) (P<0.02).

Real-time CT-fluoroscopic guided drainage is a feasible, safe, and useful therapeutic option for the management of pancreatic fistula after pancreatectomy. The fluid amylase level is a useful indicator to predict refractory pancreatic fistula.